于玻 徐珺岩 谭律岐
（北京师范大学 教育技术学院，北京 100875）
【摘 要】联通主义学习以社会网络交互、群体智慧的创造和学习者自我驱动为主要特点，知识与知识之间、学习者与学习者之间、学习者与知识之间动态聚合成学习者的认知神经网络、概念网络及社会网络。本文通过社会网络分析、话题相似度分析等方法，从社会交互维度和内容维度对cMOOC学习者进行分类和界定，通过K-means聚类分析，得到cMOOC中共有8类学习者，即沉默学习者、边缘学习者、普通学习者、专业共同体学习者、情感交互型学习者、核心学习者、桥梁学习者及导学者，并深入探究导学者及桥梁学习者的特征及在网络中的作用。每个类型的学习者都有其特点和存在的意义，并通过动态交互形成了开放、复杂、半自组织的学习生态。本研究探究出不同类型学习者的特征，期待能为个性化的学习支持服务设计提供参考。
【关键词】cMOOC；在线学习；联通主义；社会网络分析；话题相似度；学习者聚类
Clustering of cMOOC Learners Based on Social Network and Topic Similarity
YU Bo, XU Junyan and TAN lvqi
(School of Educational Technology, Beijing Normal University, Beijing 100875, China)
Abstract: Connectivism learning is characterized by interaction of social networks, the creation of group wisdom and learner self-driving. Learners’ cognitive neural networks, conceptual networks and social networks are formed dynamically between knowledge-knowledge, learners-learners and learners-knowledge. This paper classifies and defines cMOOC learners from social interaction dimension and content dimension by means of social network analysis and topic similarity analysis. Eight types of learners in cMOOC are obtained through K-means cluster analysis, namely silent learners, marginal learners, ordinary learners, professional community learners, emotional interaction learners, core learners, bridge learners and guide learners. It also explores the characteristics of bridge learners and guide learners and their roles in the network. Each type of learner has its own characteristics and significance, and forms an open, complex, semi-self-organized learning ecology through dynamic interaction. This study explores the features of different types of learners and is expected to provide references for the design of personalized learning support services.
Keywords: cMOOC; online learning; connectivity; social network analysis; topic similarity; learner clustering
下载地址：
基于社会网络和话题相似度的cMOOC学习者聚类研究.pdf
